LEADER 11030nam 2200517 450 001 996472069703316 005 20231110232438.0 010 $a3-031-04881-4 035 $a(MiAaPQ)EBC6962856 035 $a(Au-PeEL)EBL6962856 035 $a(CKB)21639818000041 035 $a(PPN)262167611 035 $a(EXLCZ)9921639818000041 100 $a20221121d2022 uy 0 101 0 $aeng 135 $aurcnu|||||||| 181 $ctxt$2rdacontent 182 $cc$2rdamedia 183 $acr$2rdacarrier 200 00$aPattern recognition and image analysis $e10th Iberian conference, IbPRIA 2022, Aveiro, Portugal, May 4-6, 2022, proceedings /$fArmando J. Pinho [and three others] 210 1$aCham, Switzerland :$cSpringer,$d[2022] 210 4$d©2022 215 $a1 online resource (704 pages) 225 1 $aLecture Notes in Computer Science ;$vv.13256 300 $aIncludes index. 311 08$aPrint version: Pinho, Armando J. Pattern Recognition and Image Analysis Cham : Springer International Publishing AG,c2022 9783031048807 327 $aIntro -- Preface -- Organization -- Plenary Talks -- The Vision Subsystems in the TrimBot2020 Gardening Robot -- Speech as Personal Identifiable Information -- Machine Learning Security: Attacks and Defenses -- Invited Tutorials -- Speech Recognition and Machine Translation: From Bayes Decision Theory to Machine Learning and Deep Neural Networks -- Human 3D Sensing from Monocular Visual Data Using Classification Techniques -- Contents -- Document Analysis -- Test Sample Selection for Handwriting Recognition Through Language Modeling -- 1 Introduction -- 2 Methodology -- 2.1 Neural End-to-end Recognition Framework -- 2.2 Language Model -- 3 Experimental Setup -- 3.1 Corpora -- 3.2 Neural Architectures -- 3.3 Best Hypothesis Selection Policies -- 3.4 Evaluation Protocol -- 4 Results -- 5 Conclusions -- References -- Classification of Untranscribed Handwritten Notarial Documents by Textual Contents -- 1 Introduction -- 2 Probabilistic Indexing of Handwritten Text Images -- 3 Plain Text Document Classification -- 3.1 Feature Selection -- 3.2 Feature Extraction -- 4 Textual-Content-Based Classification of Sets of Images -- 4.1 Estimating Text Features from Image PrIx's -- 4.2 Estimating Information Gain and TfIdf of Sets of Text Images -- 4.3 Image Document Classification -- 5 Dataset and Experimental Settings -- 5.1 A Handwritten Notarial Document Dataset -- 5.2 Empirical Settings -- 6 Experiments and Results -- 7 Conclusion -- References -- Incremental Vocabularies in Machine Translation Through Aligned Embedding Projections -- 1 Introduction -- 2 Related Work -- 3 Models -- 3.1 Tokenization -- 3.2 Transformer Architecture -- 3.3 Projecting Vectors into Different Latent Spaces -- 4 Experimental Setup -- 4.1 Datasets -- 4.2 Training Details -- 4.3 Evaluation Metrics -- 5 Experimentation -- 5.1 Projecting Pre-trained Embeddings. 327 $a5.2 On the Importance of High-Quality Embeddings -- 5.3 Zero-shot Translation -- 6 Conclusions -- 7 Future Work -- References -- An Interactive Machine Translation Framework for Modernizing the Language of Historical Documents -- 1 Introduction -- 2 Related Work -- 3 Language Modernization Approaches -- 3.1 SMT Approach -- 3.2 NMT Approaches -- 4 Interactive Machine Translation -- 4.1 Prefix-Based IMT -- 4.2 Segment-Based IMT -- 5 Experimental Framework -- 5.1 User Simulation -- 5.2 Systems -- 5.3 Corpora -- 5.4 Metrics -- 6 Results -- 6.1 Quality Analysis -- 7 Conclusions and Future Work -- References -- From Captions to Explanations:pg A Multimodal Transformer-based Architecture for Natural Language Explanation Generation -- 1 Introduction -- 2 Related Work -- 2.1 Image Captioning -- 2.2 Natural Language Explanation Generation -- 3 Proposed Methodology -- 3.1 A Synthetic Dataset to Distinguish Captions from Explanations -- 3.2 An Encoder-Decoder Vision Transformer for Natural Language Explanation Generation -- 4 Results and Discussion -- 4.1 Image Captioning -- 4.2 Natural Language Explanations -- 5 Conclusion and Future Work -- References -- Medical Image Processing -- Diagnosis of Skin Cancer Using Hierarchical Neural Networks and Metadata -- 1 Introduction -- 2 Methodology -- 2.1 Flat Classifier -- 2.2 Hierarchical Classifier -- 2.3 Methods to Combine Images and Metadata -- 2.4 Selection Between Flat and Hierarchical Models -- 3 Results -- 3.1 Dataset -- 3.2 Performance Metrics -- 3.3 Impact of Metadata on Hierarchical and Flat Models -- 3.4 Comparison of the Mixed Models -- 3.5 Elimination of Classifiers (d) and (e) -- 3.6 Evaluation on Held-out Test Set -- 4 Conclusion -- References -- Lesion-Based Chest Radiography Image Retrieval for Explainability in Pathology Detection -- 1 Introduction -- 2 Methods -- 2.1 Dataset. 327 $a2.2 CXR Pathology Object Detection -- 2.3 Lesion-Based CXR Image Retrieval -- 2.4 Structural Similarity Lesion-Based CXR Image Retrieval -- 3 Experiments -- 3.1 CXR Pathology Object Detection -- 3.2 Lesion-Based CXR Image Retrieval -- 3.3 Quantitative Evaluation -- 3.4 Qualitative Evaluation -- 4 Results -- 5 Discussion -- 6 Conclusion -- References -- Deep Learning for Diagnosis of Alzheimer's Disease with FDG-PET Neuroimaging -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 2D Slice-Level CNN Model -- 3.2 3D Subject-Level CNN Model -- 4 Experiments and Results -- 4.1 Dataset -- 4.2 2D Slice-Level CNN for AD Diagnosis with PET Data -- 4.3 3D Subject-level CNN for AD Diagnosis with PET Data -- 5 Conclusions -- References -- Deep Aesthetic Assessment and Retrieval of Breast Cancer Treatment Outcomes -- 1 Introduction -- 2 Materials and Methods -- 2.1 Data -- 2.2 Method -- 2.3 Evaluation -- 3 Results -- 4 Discussion and Conclusions -- References -- Increased Robustness in Chest X-Ray Classification Through Clinical Report-Driven Regularization -- 1 Introduction -- 2 Materials and Methods -- 2.1 Data -- 2.2 Method -- 2.3 Evaluation -- 3 Results -- 4 Discussion and Conclusions -- References -- Medical Applications -- Deep Detection Models for Measuring Epidermal Bladder Cells -- 1 Introduction -- 2 Materials and Methods -- 2.1 Computational Methods -- 2.2 Experimental Study -- 3 Results -- 4 LabelGlandula -- 5 Case Study -- 6 Conclusions and Further Work -- References -- On the Performance of Deep Learning Models for Respiratory Sound Classification Trained on Unbalanced Data -- 1 Introduction -- 2 Data Description -- 3 Experimental Setup -- 3.1 Noise Filtering -- 3.2 Input Audio Representations -- 3.3 Sampling Methods -- 3.4 Data Augmentation -- 3.5 Classifiers -- 3.6 Validation and Metrics -- 4 Results and Discussion -- 5 Conclusions. 327 $aReferences -- Automated Adequacy Assessment of Cervical Cytology Samples Using Deep Learning -- 1 Introduction -- 2 Related Work -- 3 Methodology -- 3.1 System Overview -- 3.2 Dataset -- 3.3 Experimental Setup -- 3.4 Model Evaluation -- 4 Results and Discussion -- 5 Conclusions and Future Work -- References -- Exploring Alterations in Electrocardiogram During the Postoperative Pain -- 1 Introduction -- 2 Related Literature -- 3 Methods -- 3.1 Data Collection -- 3.2 Preprocessing -- 3.3 Feature Extraction and Transformation -- 3.4 Univariate Approach -- 3.5 Unsupervised Multivariate Approach -- 4 Results -- 4.1 Dataset -- 4.2 Univariate Approach -- 4.3 Unsupervised Multivariate Approach -- 5 Discussion -- 6 Conclusions and Future Work -- References -- Differential Gene Expression Analysis of the Most Relevant Genes for Lung Cancer Prediction and Sub-type Classification -- 1 Introduction -- 2 Materials and Methods -- 2.1 Dataset and Preprocessing -- 2.2 Cancer and Subtype Classification -- 2.3 Differential Expressed Genes Analysis -- 3 Results and Discussion -- 3.1 Classification Results -- 3.2 Differential Expressed Genes Analysis -- 4 Conclusions -- References -- Detection of Epilepsy in EEGs Using Deep Sequence Models - A Comparative Study -- 1 Introduction -- 2 Methods -- 2.1 EEG Data and Pre-processing -- 2.2 Database Structuring -- 2.3 Deep Learning Models -- 2.4 Performance Assessment -- 3 Results -- 4 Discussion -- 5 Conclusion -- References -- Biometrics -- Facial Emotion Recognition for Sentiment Analysis of Social Media Data -- 1 Introduction -- 2 Related Work -- 3 Proposed Method -- 3.1 Image Classifier -- 3.2 Salient Area Detector -- 3.3 Text Classification -- 3.4 Facial Expression Recognition Module -- 3.5 Decision Fusion -- 4 Experiments -- 4.1 Training the Facial Expression Recognition Model -- 4.2 Data Set for the Full Model Evaluation. 327 $a4.3 Results -- 5 Conclusions -- References -- Heartbeat Selection Based on Outlier Removal -- 1 Introduction -- 2 Proposed Approach -- 2.1 Modified DMEAN: Formal Description -- 2.2 Algorithm Tunning: Feature and Threshold Selection -- 2.3 Normalized Cross-Correlation -- 3 Methodology -- 3.1 Pre-processing -- 3.2 Modified DMEAN in ECG Biometrics -- 4 Performance Analysis -- 4.1 Performance Analysis Between Modified DMEAN and DMEAN -- 4.2 Performance Analysis of Modified DMEAN and NCC -- 5 Conclusions -- References -- Characterization of Emotions Through Facial Electromyogram Signals -- 1 Introduction -- 2 Dataset and Methodology -- 2.1 Dataset -- 2.2 Methodology -- 3 Results -- 4 Conclusion -- References -- Multimodal Feature Evaluation and Fusion for Emotional Well-Being Monitorization -- 1 Introduction -- 2 Task and Corpus -- 3 Feature Analysis -- 3.1 Semantic and Paralinguistic Information -- 3.2 Clustering -- 3.3 Novel Techniques for Semantic Information Extraction -- 4 Supervised Learning for Classification Experiments -- 4.1 Random Forest -- 4.2 ANN -- 4.3 BERT -- 4.4 Results -- 5 Conclusions and Future Work -- References -- Temporal Convolutional Networks for Robust Face Liveness Detection -- 1 Introduction -- 2 Related Work -- 2.1 Non-intrusive Pulse Extraction -- 2.2 Face Liveness Detection -- 3 Robust Liveness Detection on the Fly -- 4 Temporal Convolutional Neural Networks for Liveness Detection -- 4.1 Regular Convolution TCN Block -- 4.2 Dilated Convolution TCN Block -- 5 Evaluation -- 5.1 Dataset -- 5.2 Protocol and Implementation Details -- 5.3 Results -- 5.4 TCN Model Training Analysis -- 6 Conclusions -- References -- Pattern Recognition and Machine Learning -- MaxDropoutV2: An Improved Method to Drop Out Neurons in Convolutional Neural Networks -- 1 Introduction -- 2 Related Works -- 3 MaxDropoutV2 as an Improved Version of MaxDropout. 327 $a3.1 MaxDropout. 410 0$aLecture Notes in Computer Science 606 $aImage processing$xDigital techniques 606 $aOptical pattern recognition 615 0$aImage processing$xDigital techniques. 615 0$aOptical pattern recognition. 676 $a006.4 702 $aPinho$b Armando J. 801 0$bMiAaPQ 801 1$bMiAaPQ 801 2$bMiAaPQ 906 $aBOOK 912 $a996472069703316 996 $aPattern Recognition and Image Analysis$92838914 997 $aUNISA LEADER 03962nam 2200589 450 001 9910811409503321 005 20230808200629.0 010 $a1-5017-5758-X 010 $a1-60909-204-X 024 7 $a10.1515/9781501757587 035 $a(CKB)3710000000957118 035 $a(MiAaPQ)EBC4745800 035 $a(OCoLC)953387299 035 $a(MdBmJHUP)muse57192 035 $a(DE-B1597)572391 035 $a(DE-B1597)9781501757587 035 $a(Au-PeEL)EBL4745800 035 $a(CaPaEBR)ebr11300706 035 $a(OCoLC)964283061 035 $a(OCoLC)1229161680 035 $a(EXLCZ)993710000000957118 100 $a20161212h20162016 uy 0 101 0 $aeng 135 $aurcnu|||||||| 181 $2rdacontent 182 $2rdamedia 183 $2rdacarrier 200 10$aSocialist churches $eradical secularization and the preservation of the past in Petrograd and Leningrad, 1918-1988 /$fCatriona Kelly 210 1$aDeKalb, Illinois :$cNorthern Illinois University Press,$d2016. 210 4$d©2016 215 $a1 online resource (433 pages) $cillustrations, photographs 225 0 $aNIU Series in Slavic, East European, and Eurasian Studies 311 $a0-87580-743-7 320 $aIncludes bibliographical references and index. 327 $a"October has caught up with the church": the separation of church and state, 1918-1923 -- Monuments to the golden age: the canons of preservation, 1924-1928 -- Churches in the Socialist city: crash industrialization, rational Atheism, and city planning, 1929-1940 -- The great patriotic church war destruction, post-war reconstruction, 1941-1953 -- The scientific assault on God: church-monuments in the Khrushchev era, 1953-1964 -- Cynosures of the city: church buildings as national heritage, 1965-1988. 330 $aIn Russia, legislation on the separation of church and state in early 1918 marginalized religious faith and raised pressing questions about what was to be done with church buildings. While associated with suspect beliefs, they were also regarded as structures with potential practical uses, and some were considered works of art. This engaging study draws on religious anthropology, sociology, cultural studies, and history to explore the fate of these "socialist churches," showing how attitudes and practices related to them were shaped both by laws on the preservation of monuments and anti-religious measures. Advocates of preservation, while sincere in their desire to save the buildings, were indifferent, if not hostile, to their religious purpose. Believers, on the other hand, regarded preservation laws as irritants, except when they provided leverage for use of the buildings by church communities. The situation was eased by the growing rapprochement of the Orthodox Church and Soviet state organizations after 1943, but not fully resolved until the Soviet Union fell apart. Based on abundant archival documentation, Catriona Kelly's powerful narrative portrays the human tragedies and compromises, but also the remarkable achievements, of those who fought to preserve these important buildings over the course of seven decades of state atheism. Socialist Churches will appeal to specialists, students, and general readers interested in church history, the history of architecture, and Russian art, history, and cultural studies. 606 $aAtheism$zSoviet Union 606 $aChurch buildings$zSoviet Union 606 $aHistoric buildings$xConservation and restoration$zSoviet Union 610 $achurches as works of art, Russian churches, church architecture. 615 0$aAtheism 615 0$aChurch buildings 615 0$aHistoric buildings$xConservation and restoration 676 $a322/.109470904 700 $aKelly$b Catriona$0505856 801 0$bMiAaPQ 801 1$bMiAaPQ 801 2$bMiAaPQ 906 $aBOOK 912 $a9910811409503321 996 $aSocialist churches$94037706 997 $aUNINA